The replacement of ageing infrastructure is not sustainable, not cost effective, not convenient, not safe, and sometimes not possible
Worldwide, the need for strategic infrastructure requires an annual expenditure of more than US$3.7 trillion, or 5% of global GDP. Since the current supply is growing at only $2.7 trillion each year, there is an infrastructure spending gap that is increasing by US$1 trillion per year. This gap can be reduced with more informed decisions related to ageing infrastructure.
Civil infrastructure often has much reserve capacity since behaviour models at the design-stage, prior to construction, are necessarily conservative. More accurate estimates of real behaviour are needed once infrastructure is in service. Improved knowledge of current performance leads to better predictions of performance when weighing decisions such as extension, improvement, repair and replacement.
We use sensor measurement data combined with site-inspection results and engineering knowledge to improve behaviour models. Inspired by fundamental research in model-based diagnosis, we are developing sensor-data-interpretation methodologies for full-scale applications where complete knowledge of uncertainties is not available. Current strategies for interpreting data are weak. Such strategies do not adequately account for the typically high levels of systematic modelling uncertainty and when uncertainty information is not complete, interpretation – and most critically, predictions – may be biased.
The implementation of cyber civil infrastructure has the potential to ensure that massive investments are optimally managed, that infrastructure modifications are well engineered, and that future designs are improved. Building on experience of more than 15 years of civil-engineering research and development – including several full-scale applications in four countries in Europe and North America – this project focuses on the development of robust methodologies that use sensor data to improve decision making.